An IoT-based System for Air Pollution Data Analysis and Visualization

Air quality has become the most concerning matter around the world. The World air quality report 2021 claimed that South Asia, including Bangladesh, is the worst air polluted country, where Dhaka is the second most polluted city. Air pollution causes everybody to see the dilemma as it causes severe diseases related to the lungs, which can be acute or chronic. Good air quality depends on the particular amount of different substances within a range in the air; otherwise, it becomes a matter of concern. These measurements rely on various ingredients such as the rate of CO, CO2, NO2, Alcohol, Benzene, Particulate Matter, Dust, and so on. Our primary purpose is to build an IoT-based model to collect data from the sensors. We have used five parameters for monitoring pollution. These are the smoke, CO, particulate matter, temperature, and humidity. A real-time monitoring system is developed for the front-end that allows knowing the current situation. This work also includes the data analysis, which predicts the condition according to the previous data with a machine-learning model. A comparison of various prediction algorithms, including performance measures and test case analysis, is presented. We also analyze the correlation between Air Quality Index (AQI) and weather conditions (temperature and humidity) for different locations and times.